upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-23 11:10:55 -05:00
Code Issues Packages Projects Releases Wiki Activity

model: add Mellum architecture (#23966)

Browse Source 
* model: support for Mellum architecture

* model: improve mellum.py formatting

* model: improve mellum.py formatting once again

* deps: downgrade transformers to 4.57.6 (to fix CI)

* deps: remove huggingface_hub dependency

* deps: remove huggingface_hub from test requirements

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Mikhail Podvitskii
2026-06-02 21:11:12 +02:00
committed by GitHub
parent bfb4308b05
commit 4fb16eccce
20 changed files with 344 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
gguf-py/gguf/constants.py
View File

@@ -510,6 +510,7 @@ class MODEL_ARCH(IntEnum):
MAINCODER = auto()
KIMI_LINEAR = auto()
TALKIE = auto()
MELLUM = auto()
class VISION_PROJECTOR_TYPE(IntEnum):
@@ -1030,6 +1031,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
MODEL_ARCH.MAINCODER: "maincoder",
MODEL_ARCH.KIMI_LINEAR: "kimi-linear",
MODEL_ARCH.TALKIE: "talkie",
MODEL_ARCH.MELLUM: "mellum",
}
VISION_PROJECTOR_TYPE_NAMES: dict[VISION_PROJECTOR_TYPE, str] = {
@@ -4093,6 +4095,23 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.FFN_UP,
MODEL_TENSOR.LAYER_OUT_SCALE,
],
MODEL_ARCH.MELLUM: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_Q_NORM,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_K_NORM,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_GATE_INP,
MODEL_TENSOR.FFN_GATE_EXP,
MODEL_TENSOR.FFN_DOWN_EXP,
MODEL_TENSOR.FFN_UP_EXP,
],
# TODO
}